Anomaly detection for medical samples under multiple settings

Abstract

Using multiple imaging modes in whole slide image screening is potentially useful to reduce false positives. To use multiple imaging modes, a method for locating anomalies on a medical sample from an image thereof uses an anomaly-detection process that comprises using plural base classifiers individually to classify an object-of-interest suspected to be an anomaly. Each base classifier respectively extracts features of the object-of-interest and generates, according to the extracted features, a score indicating a likelihood of the object-of-interest being anomalous. The anomaly-detection process further comprises using an aggregate classifier to combine the scores generated by the base classifiers to determine whether the object-of-interest is the anomaly. The aggregate classifier determines a dependability measure for each base classifier according to setting-based variables of a setting under which the sample and the image are obtained, and then selectively combines the scores of the base classifiers according to the dependability measures.

Description

FIELD OF THE INVENTION[0001]The present invention relates to anomaly detection for a medical sample in computer aided diagnosis (CAD).BACKGROUND[0002]Anomaly detection is crucial for CAD. Whole slide image (WSI) searching is preferred for CAD to avoid false negatives but is limited by its image acquisition and processing speed. Most of the existing approaches for WSI screening use low-magnification images in order to reduce image acquisition time. Thus, image quality becomes degraded. Information contained in low-magnification images is less than that in high-resolution ones, thus imposing limitations in accurately segmenting and classifying anomalies. Moreover, large searching area in WSI screening introduces more false positives, which affect diagnostic results as well.[0003]Using multiple imaging modes in WSI screening is potentially useful to reduce false positives. At present, most products do not provide WSI analysis for evaluation and do not support multiple imaging modes. Th...

AI Technical Summary

Benefits of technology

Improves anomaly detection accuracy by intelligently combining results from different classifiers based on experimental and imaging conditions, effectively reducing false positives and enhancing diagnostic precision even with low-magnification images.

Problems solved by technology

Whole slide image (WSI) searching is preferred for CAD to avoid false negatives but is limited by its image acquisition and processing speed.

Thus, image quality becomes degraded.

Information contained in low-magnification images is less than that in high-resolution ones, thus imposing limitations in accurately segmenting and classifying anomalies.

Moreover, large searching area in WSI screening introduces more false positives, which affect diagnostic results as well.

At present, most products do not provide WSI analysis for evaluation and do not support multiple imaging modes.

Besides, most of them are machine dependent, which are not easy to be adapted to new hardware settings.

It may cause false positive results, thereby affecting accuracy of diagnostic results.

Images